The purpose of this project is: a) to develop models that explain how psychiatric hospitals make staffing decisions and how those decisions affect patient outcomes; b) to estimate and test those models; and c) to use the best ones to aid in policy analysis. The study will: (1) evaluate the influence of patient, hospital, and community characteristics, and nurse supply on a hospital's employment of psychiatric nursing labor; (2) evaluate the effect of a hospital's choice of staffing and skill-mix levels on patient outcomes; and (3) compare the supply function of psychiatric nurses with the population of registered nurses. Finally, the study will evaluate the contribution of classical economic and organizational theories to the understanding of the psychiatric hospital's use of labor. An organization level evaluation will be based on secondary data from the Quality of Care/Medpar file (HCFA), an Inventory of Mental Health Organizations (NIMH), a Client Sample Survey of Mental Health Organizations (NIMH), Medicare Cost Report data (HCFA), the Area Resources File, and the American Hospital Association's Annual Survey of Hospitals. Patient outcomes will be measured by number of AMA discharges, length of stay, percentage of readmissions within 30 days, and cost of care. In addition, a primary data collection of supply characteristics will be undertaken. This survey will measure current demographic and employment characteristics, work activities, salary, career patterns, populations served, job satisfaction, perceptions of quality of care, and perceptions of support staff of psychiatric nurses. This study will provide knowledge to the industry on the supply characteristics of psychiatric nurses; it will evaluate the effect on patient outcomes of the selection of the staffing and skill-mix levels, of a hospital's labor force. Impact analyses and simulations will be done to evaluate the effect of changing parameter values based on a range of related policy decisions. Policy issues to be evaluated include: effect of substitution of different types of professional and non-professional labor for psychiatric nurses; effect of changes in patient length of stay, case-mix, and treatments on nurse labor use and outcomes; and effect of different skill-mix levels of RNs on patient outcomes.